Research On Channel Estimation Error Based Channel Reciprocity Compensation Algorithm In Massive MIMO

In recent years,with the deepening of people's research on 5G technology,massive MIMO has become one of the key technologies of the future 5G technologies.The two duplex modes: Time Division Duplex(also called TDD)and Frequency Division Duplex(FDD),can be applied to wireless communications systems,TDD is widely used in massive MIMO systems due to its reciprocity of uplink and downlink channels.During the actual transmission of mobile communications,however,channel reciprocity is severely damaged due to its transmission environment and other aspects,so some effective compensation schemes must be designed to ensure that massive MIMO is not affected by these aspects.In this thesis,some reciprocity algorithms to compensate for channel estimation error are proposed.The main work of this thesis is as follows:Firstly,the advantages of massive MIMO over traditional MIMO systems,and the working principle,characteristics and development trend of OFDM systems are briefly introduced.Then,the advantages of TDD systems are analyzed and factors that results in the loss of channel reciprocity are discussed,and briefly described the corresponding compensation algorithm.Secondly,the influence of channel estimation error on massive MIMO systems is analyzed,and it is proved that estimation error can undermine reciprocity in TDD systems and lead to a decrease in channel performance.On such basis,this thesis proposes an algorithm based on MMSE criterion.The algorithm predicts downlink channel information by calibrating the uplink channel state information,and calculates the calibration matrix by using MMSE criterion.The simulations show that the algorithm will effectively reduce the impact of channel estimation error and improve system performance.Considering that the above algorithm needs to use channel statistical information,autocorrelation matrix,which is not not known perfectly,the thesis propose an iterative reciprocity compensation algorithm.The accurate autocorrelation matrix and calibration matrix are obtained after some iteration,so as to improve the accuracy of the downlink channel prediction.Finally,the thesis focuses on the influence of the channel estimation error detected by base station and terminals,a DFP based compensation algorithm is proposed.The algorithm constructs a Rayleigh quotient and uses DFP(Davidon-Fletcher-Powell)algorithm to estimate calibration matrix.The analysis and simulation show that the algorithm can effectively improve the system performance and has lower complexity.